|
Mojtaba Safari Postdoctoral Scholar · University of Chicago I'm a postdoctoral scholar in the Department of Radiation and Cellular Oncology at the University of Chicago, where I am fortunate to work with Xiaofeng Yang. My research develops physics-informed AI for MRI reconstruction, super-resolution, motion correction, and quantitative imaging, with the goal of making image-guided radiation therapy faster and more accurate. Most of this work appears in medical physics and medical imaging journals, and I review for several of them. I did my PhD in physics at Université Laval, advised by Louis Archambault and Ali Fatemi, with a thesis on Using MRI Physics and AI for Multi-Parametric MRI-Guided Radiotherapy. Before that I studied at IMT Atlantique and Sharif University of Technology. |
|
ResearchI'm interested in medical physics, MRI physics, medical image analysis, and machine learning for medical imaging. Most of my work builds physics-informed models that make MRI faster, cleaner, and more quantitative, then puts those images to use in radiation therapy. Some papers are highlighted. |
Research Collaboration |
Research Interests |
I am interested in collaborative research that advances medical imaging,
medical physics, and artificial intelligence for healthcare. My current
research interests include:
• Physics-informed MRI reconstruction and accelerated imaging • Motion detection and correction in MRI • Super-resolution and quantitative MRI • MRI-guided and adaptive radiation therapy • Medical image synthesis, including synthetic CT generation • Radiomics and imaging biomarkers • Foundation and vision-language models for medical imaging I welcome discussions with researchers, clinicians, and students whose interests align with these areas. University of Chicago students are especially encouraged to reach out to discuss shared research interests and ongoing work. Students interested in formal research opportunities within our group are encouraged to contact Prof. Xiaofeng Yang. I also welcome inquiries from researchers and students at other institutions interested in scientific collaboration. |
Selected Journal ArticlesA selection. My complete and current publication list is on Google Scholar. |
-
Motion Artifact-Aware Self-Supervised Representation Learning for 3D Brain MRI Motion Artifact Reduction
IEEE Transactions on Medical Imagingdoibib
A motion artifact-aware unpaired representation learning framework for MoCo without using paired training data
-
Foundation Models in Medical Image Analysis: A Systematic Review and Quantitative Analysis
A review and meta-analysis of current medical foundation models.
-
Low-dose CT imaging using a regularization-enhanced efficient diffusion probabilistic model
A regularization-enhanced diffusion probabilistic model for low-dose CT reconstruction.
-
A digital twin framework for adaptive treatment planning in radiotherapy
Physics in Medicine & Biologydoibib
A digital twin framework that adapts radiotherapy treatment plans online.
-
A generalist biomedical vision-language model via multi-CLIP knowledge distillation
Nature Communicationsdoicodebib
A generalist biomedical vision-language model distilled from multiple CLIP teachers.
-
Efficient vision mamba for MRI super-resolution via hybrid selective scanning
A hybrid selective-scanning vision Mamba architecture for MRI super-resolution.
-
Transformer-based synthesis of 7 Tesla-like T1 contrast from routine clinical magnetic resonance imaging
A transformer that synthesizes 7 Tesla-like T1 contrast from routine clinical MRI.
-
Vision foundation model for 3D magnetic resonance imaging segmentation, classification, and registration
Medical Image Analysisdoicodebib
A 3D MRI foundation model evaluated across segmentation, classification, and registration.
-
Reinforcement learning-guided segment anything model for MRI prostate and dominant intraprostatic lesions auto-segmentation
Physics in Medicine & Biologydoibib
Reinforcement learning guides SAM to segment the prostate and dominant intraprostatic lesions.
-
An efficient 3D latent diffusion model for T1-contrast enhanced MRI generation
Biomedical Physics & Engineering Expressdoibib
An efficient 3D latent diffusion model for generating T1-contrast enhanced MRI.
-
Systematic review and meta-analysis of AI-driven MRI motion artifact detection and correction
A systematic review and meta-analysis of AI methods for detecting and correcting MRI motion artifacts.
-
Advancing MRI reconstruction: A systematic review of deep learning and compressed sensing integration
Biomedical Signal Processing and Controldoicodebib
A systematic review of how deep learning and compressed sensing are combined for MRI reconstruction.
-
A physics-informed deep learning model for MRI brain motion correction
A physics-informed deep learning model for correcting motion artifacts in brain MRI.
-
Res-MoCoDiff: residual-guided diffusion models for motion artifact correction in brain MRI
Physics in Medicine & Biologydoibib
A residual-guided diffusion model for correcting motion artifacts in brain MRI.
-
Foundation model based multimodal transformer framework for survival analysis in HER2 stratified breast cancer
Physics in Medicine & Biologydoibib
A foundation-model transformer for survival analysis in HER2-stratified breast cancer.
-
MRI super-resolution reconstruction using efficient diffusion probabilistic model with residual shifting
Physics in Medicine & Biologydoicodebib
An efficient diffusion probabilistic model with residual shifting for MRI super-resolution.
-
Self-supervised adversarial diffusion models for fast MRI reconstruction
A self-supervised adversarial diffusion approach to accelerating MRI reconstruction.
-
T1-contrast enhanced MRI generation from multi-parametric MRI for glioma patients with latent tumor conditioning
Latent tumor conditioning for generating T1-contrast enhanced MRI in glioma patients.
-
Deep learning based apparent diffusion coefficient map generation from multi-parametric MR images for patients with diffuse gliomas
Deep learning generates apparent diffusion coefficient maps from multi-parametric MRI.
-
Unsupervised MRI motion artifact disentanglement: introducing MAUDGAN
Physics in Medicine & Biologydoibib
An unsupervised GAN that disentangles motion artifacts from the underlying MRI anatomy.
-
MRI motion artifact reduction using a conditional diffusion probabilistic model (MAR-CDPM)
Selected among the top 10% most-viewed articles in Medical Physics, 2023
A conditional diffusion probabilistic model for reducing MRI motion artifacts.
-
MedFusionGAN: multimodal medical image fusion using an unsupervised deep generative adversarial network
An unsupervised generative adversarial network for fusing multimodal medical images.
Selected Conference Papers & AbstractsA selection, weighted towards work I presented myself. |
-
Text-Guided Segmentation of Glioma Substructures on Multi-Parametric MRI
AAPM Annual Meeting, Vancouver, BCbib
Best in Physics (Multi-Disciplinary) Award
-
Efficient Mamba network for super-resolution prostate MR imaging
SPIE Medical Imaging, Vancouver, BCabstractbib
Imaging Informatics Best Paper Award and Imaging Informatics Cum Laude Poster Award
-
A Multimodal Foundation Model for Pediatric Multi-parametric MRI Synthesis
AAPM Annual Meeting, Vancouver, BCbib
Selected for the Science Council Session on Innovations in Medical Physics, one of 12 abstracts chosen from over 250 submissions
-
Performance of GPT-5 in brain tumor MRI reasoning
Proc. SPIE 13929, Medical Imaging 2026: Clinical and Biomedical Imaging, Vancouver, BCdoiabstractbib
-
Capabilities of GPT-5 on multimodal medical reasoning
Proc. SPIE 13930, Medical Imaging 2026: Imaging Informatics, Vancouver, BCdoicodebib
-
Adversarial diffusion-based self-supervised learning for high-resolution MR imaging
AAPM Annual Meeting, Washington, DCbib
-
A vision-language deep learning model for predicting survival outcomes in glioblastoma patients
AAPM Annual Meeting, Washington, DCbib
-
A physics-informed deep learning method for correcting motion artifacts in brain MR imaging
ISMRM, Honolulu, HIbib
-
Text-enhanced vision-language motion correction (VLM-MoCo) for mitigating severe motion artifacts in MRI scans
ISMRM, Honolulu, HIbib
Academic Service |
Reviewing |
Medical Physics, 2021–present
Physics in Medicine and Biology and Machine Learning: Science and Technology, 2024–present (IOP Trusted Reviewer, 2025) Medical Image Analysis, 2025–present IEEE Transactions on Medical Imaging, 2025–present IEEE Transactions on Image Processing, 2024–present IEEE Transactions on Biomedical Engineering, 2026–present Biomedical Signal Processing and Control, 2024–present Med (Cell Press), 2026–present |
|
|